• State
  • Locked Locked
  • Replies 1 reply
  • Subscribers 79 subscribers
  • Views 379 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

Let's Talk ADC10

Zack Bomsta
Intellectual 470 points
Other Parts Discussed in Thread: MSP430F2274

Hi All,

Just a few questions about the ADC10 on the MSP430F2274. Recently, I have been using modules that have the ADC12 on it. As a result, I have gotten use to the flexibility and power of the ADC12. Recently, I had to use the MSP430F2274 again in a design, and unfortunately, I designed my PCB with the flexibility of the ADC12 in mind. More specifically, I did not realize that the "Repeat-Sequence-of-Channels Mode" on the ADC10 decrements from INCHx to INCH0 without the ability to select the order and number of channels sampled in the sequence like on the ADC12. My current application is battery powered and needs to be as low-power as possible. I need to sample 3 analog signals every 15 seconds or so, but unfortunately, as Murphy's Law would have it, I placed one of the analog signals on A15, and another one on A0. I don't want to have to leave the ADC on while it performs 13 channels that I don't need, and I would like to minimize CPU intervention. Any suggestions on a work-around to keep the current consumption to a minimum? Also, what happens when the ADC samples a channel that has not had the analog input enabled (i.e. several of the ports between A15 and A0 are being used for digital functions. How will the ADC treat them?)

Finally, I have always added an anti-aliasing filter on the input of each ADC channel to help improve signal integrity. The filter is usually a basic low pass with a 10K series resistor and a .1uF cap to ground. After reading the user's manual more closely, it looks like the analog inputs are shorted to ground when not being sampled. If that is the case, that would mean that the cap on my low pass filter would greatly increase the required sample time of the ADC since the cap has to recharge right? Should I take it off and let the natural low pass filter of the ADC do the filtering for me? Does anyone have any advice here?

Thanks for all your help. Sorry this post was so long.

  • 0
    Guru 227245 points

    Zack Bomsta said:
    Any suggestions on a work-around to keep the current consumption to a minimum?

    ALl you can do is doing individual signle-channel conversions and when the 'ready' interrupt comes, wake the CPU and start the next conversion.

    Zack Bomsta said:
    what happens when the ADC samples a channel that has not had the analog input enabled

    If you look at the por tpin schematics in teh datasheet, you'll see that (at least for the very most devices) analog funcitonality is always enabled. The 'enable' bits actually disable the digital part so that it has no incluence on the input. If you don't do it, you are sampling e.g. the output voltage of the port output driver (which can be used to estimate the load current based on the observed voltage drop)

    Zack Bomsta said:
    After reading the user's manual more closely, it looks like the analog inputs are shorted to ground when not being sampled

    No, the analog multiplexer is a break-before-make T-switch. That measn there are three analog switches. If a channel is currently sampled (the sampling capacitor is charged) then switch 1 and 3 are closed and switch 1 is open, connecting the input with the sampling capacitor. When sampling is done and conversion starts (or channel is not selected) then first the two switches 1+3 open, then switch 2 closes, shorting their middle point to GND. This minimizes crosstalk between the channels. However at no time, input or sampling capacitor are shorted to GND.

    If you want the 160Hz low-pass, then keep it. It even decreases sampling time (and therefore the ADC on time), as it acts as an impedance changer. Compare dwith your signal source, the capacitor is relatively low impedance, so the sampling time can be shorter than it could be when there were only the (assumed high-impedance) source attached.

    However, the filter capacitor has some leakage current which may influence the precision as it forms a voltage divider with the series resistor and the source impedance. so don't use a cheap, leaky one. :)

**Attention** This is a public forum